Senior AI Infrastructure Engineer, Kubernetes

Firmus Technologies•San Francisco, CA

10d•Onsite

About The Position

Firmus Technologies is a global leader in pioneering the development and operation of efficient AI infrastructure across Asia Pacific. Founded in Australia in 2019, their mission is to create the most efficient AI infrastructure by combining cutting-edge technology with a steadfast commitment to sustainability. Firmus designs, builds, and operates a new class of digital infrastructure – the AI Factory. Through their model-to-grid technology approach, they have pushed the boundaries of multi-generational liquid cooling systems, energy management, AI software orchestration, and construction. Their large-scale GPU cloud platform, Firmus AI Cloud, is purpose-built to deliver energy-efficient AI compute at scale to customers, empowering developers, enterprises, educational institutions, and government users to train and deploy AI models with unmatched efficiency and cost savings. The Senior Kubernetes Engineer, AI Infrastructure owns the technical design and delivery of the backend infrastructure that powers the Firmus Kubernetes platform. This is a hands-on principal-level individual contributor role, responsible for building production-grade cluster lifecycle, control-plane, networking, storage, security, observability, and automation capabilities across GPU-accelerated bare-metal environments. They solve the hardest platform engineering problems, set Kubernetes engineering standards, and provide domain-level technical sign-off for platform designs. They work across AI Platforms, Solutions Architecture & Delivery, networking, security, and operations to create a secure, resilient, multi-tenant platform that can be deployed and operated consistently at AI-factory scale.

Requirements

7+ years of progressive infrastructure, systems, or platform engineering experience, including substantial ownership of production Kubernetes platforms and at least 3 years operating at senior staff, principal, or equivalent level.
Deep knowledge of Kubernetes internals, including the API server, etcd, scheduler, controller manager, kubelet, admission, CRI, CNI, CSI, reconciliation patterns, cluster performance, upgrades, and control-plane failure modes.
Demonstrated experience designing, building, and operating highly available, large scale and multi-cluster Kubernetes platforms on bare metal, private cloud, or hybrid infrastructure.
Strong software engineering ability in Go and/or Rust, with practical Python and Bash skills; experience building Kubernetes operators, controllers, admission webhooks, CLIs, or platform services.
Expert Linux systems knowledge, including namespaces, cgroups, systemd, kernel, host networking and container runtime behaviour, performance analysis, and low-level troubleshooting.
Strong Kubernetes networking expertise across Cilium, Calico, or equivalent CNI implementations, plus load balancing, DNS, ingress, BGP, network policy, and multi-network architectures.
Strong infrastructure automation and GitOps experience with tools such as Terraform, Ansible, Argo CD, Flux, GitHub Actions, GitLab CI, or Jenkins.
Practical experience with Kubernetes security and governance, including RBAC, OPA Gatekeeper or Kyverno, secrets management, certificate lifecycle, image security, and workload isolation.
Experience implementing production observability with Prometheus, Grafana, OpenTelemetry, Loki, Elasticsearch, or equivalent technologies, and using telemetry to manage reliability, capacity, and performance.
Experience with GPU-enabled Kubernetes infrastructure, NVIDIA GPU Operator, accelerator scheduling for AI workloads at large scale, RDMA networking, and distributed AI workload requirements.
Experience with distributed storage and data services such as Ceph, CSI-backed storage, object storage, backup and restore, and disaster recovery.
CKA-level expertise is expected; CKA, CKS, or relevant cloud-native certifications are strongly preferred.
Bachelor’s degree in computer science, engineering, or a related discipline, or equivalent depth of practical engineering experience.
Clear technical judgement and communication, with a record of influencing architecture across software, networking, security, platform, and operations teams.

Responsibilities

Define and own the Kubernetes platform reference architecture across management and workload clusters, including control-plane topology, cluster lifecycle, multi-tenancy, workload isolation, and failure-domain design.
Build and maintain the backend services, APIs, controllers, operators, and automation required to provision, configure, upgrade, scale, and retire Kubernetes clusters reliably.
Engineer repeatable bare-metal Kubernetes deployment and lifecycle workflows using infrastructure-as-code and automated provisioning technologies such as Cluster API, kubeadm, Redfish, PXE, Ironic, or Metal3.
Design and operate cluster networking across CNI, ingress, service discovery, DNS, load balancing, network policy, and service mesh; integrate Multus, SR-IOV, BGP, InfiniBand, or RoCE where required for high-performance AI workloads.
Define persistent-storage and data-service patterns using CSI, Ceph, local NVMe, object storage, backup and restore, and disaster-recovery mechanisms appropriate for stateful platform and AI workloads.
Integrate and productionise NVIDIA GPU and Network Operators, device plugins, drivers, DCGM telemetry, scheduling, quotas, and topology-aware placement for multi-node accelerated workloads.
Establish GitOps and CI/CD patterns for platform software, configuration, policy, and release management, with safe testing, progressive rollout, rollback, and upgrade practices.
Build platform security into the architecture through identity and access control, RBAC, secrets management, policy-as-code, image and software-supply-chain controls, tenant isolation, and auditable change management.
Define service-level objectives and engineer observability for metrics, logs, traces, events, capacity, and performance; lead diagnosis of complex distributed systems failures and eliminate recurring operational toil.
Set engineering standards, design patterns, review practices, and operational readiness criteria; mentor senior engineers and resolve cross-team technical decisions while remaining directly involved in implementation.